From one recruiter to hundreds.

Decide what you record and how it hangs together: objects, fields and the relations between them are yours to set, not a fixed shape you have to work around. That includes what a record looks like on screen: which blocks it shows and in what order.

Per field you decide: required, so a record cannot be saved without it. Unique, so the same value cannot be entered twice. Searchable, so it can be filtered and sorted on. AI-extractable, so the AI may read it out of a CV or a conversation. Sensitive, so it stays hidden until someone is explicitly given access.

A date field only accepts a date, an email field only an email address, a number only a number. Bad data is refused at the point of entry rather than found later.

Build a list the way you want to see it, with your filters, your columns and your sort order. Save it and open it again tomorrow. Share it with the team or keep it to yourself.

Candidates, jobs, applications, companies and interviews, and the relations between them.

Including recruiter feedback per candidate, so a decision keeps its reasoning.

Not one pipeline, as many as you want, and not only for candidates: build them for contacts, companies and jobs too, each with its own stages and its own look. You can even tie a pipeline to a specific client, so a job at that company automatically runs through the process you agreed with them rather than your default one. Every record shows where it stands and the history of every step it took to get there.

A skill carries a level, years of experience and its synonyms. ‘JS’ and ‘JavaScript’ are the same thing, and three years is not five.

A chat that knows your workspace: your data model, your records, your conversations. Ask it who you spoke to about a client and it finds the records. Ask it to add a field or restructure an object and it prepares the change. Anything it wants to write comes back as a proposal first.

The AI may read anything you can read, and it may prepare a change, but it cannot save one. Every change it wants to make arrives as a proposal: shown next to what is there now, waiting for you to accept, edit or throw it away. That holds for a record, for a field, for the data model itself, and for anything an automation prepares. There is no setting that turns it off by accident.

Connect an API key from whichever AI provider you already use, OpenAI, Anthropic, Azure or another, and Simply runs on that. Your provider bills you directly for what you use, at their price. Simply is never in that payment path, so there is no margin to add and no credits to buy. Switch provider and you swap one key.

Every call is logged with the model it used and what it cost, in euros, down to the individual action. So you can see that a CV took two cents and a long summary took eight, work out what a busy month actually runs to, and spot it the day something starts costing more than it should. No estimate, no monthly surprise.

Type what you are looking for the way you would say it, like “a PHP developer with ten years of experience in Utrecht”, and Simply sets the filters for you. It searches on meaning as well as on the words, so a candidate who only ever wrote “Laravel” still turns up for PHP, and the closest fit ranks first.

Score every candidate against one job, or one candidate against every open job you have. The same engine runs both directions and writes one report per pair. Each match carries a score, the reasons behind it, which requirements are not met, and how the score was reached, so you can argue with it rather than take it on faith.

Start an enrichment run on a candidate or contact and Simply goes looking for the details you are missing, like a phone number or an email address, and proposes them on the record for you to accept.

Start typing a company and pick it from the list. Simply holds a large pool of companies, kept warm across everyone who uses it, and fills the record from it: industry, size, year founded, head office, LinkedIn page, a description, even the logo, colours and fonts. You get a client record that is complete before you have typed a second field, and it stays current without anyone maintaining it.

Email, WhatsApp and LinkedIn all in the same inbox.

Whatever channel a message arrives on, it joins the same thread, and Simply attaches that thread to the right candidate or contact itself, by email address, phone number or LinkedIn profile. So a person has one history with you rather than four half-histories in four tools.

Labels every reply as it lands: interested, not interested, a question, an objection, out of office, wrong person. So you start on the promising conversations instead of working down the list. This is a classification model rather than a language model. It runs in-process on our own infrastructure, so an ordinary reply is never sent anywhere to be read. Only when the model is genuinely unsure does it escalate to a language model for a second opinion.

Simply writes a reply in the tone of the channel it is going out on, because a LinkedIn message does not read like an email, grounded in what was actually said earlier in that conversation. You approve it, edit it, or throw it away. Nothing is ever sent on its own.

Someone drops a phone number into a LinkedIn message and the record does not have it yet. Simply spots that, and offers to add it right there in the conversation. One click and it is on the candidate, so you can WhatsApp them without going to look for it. The same for an email address, a new employer, a notice period. It is a proposal like any other: nothing lands on the record until you say yes.

Keep the messages you send over and over as templates and fire one off with a single click, filled in with the candidate and the job it is going out about.

Reach out first, on any connected channel, without leaving the inbox.

When someone asks to meet, Simply proposes times in the same conversation, and you confirm.

Connect your VoIP and Simply records the call. Through a unique partnership with Odido that includes mobile numbers, inbound as well as outbound. That second half is the one most tools drop.

Google Meet and Microsoft Teams natively, with no bot joining the call. A deep integration with Google and Microsoft hands us the transcript directly, so nobody sees a stranger in the participant list.

Our iOS and Android app records the conversation at the table and drops it into the record like any other channel.

All three channels get the same treatment: a phone call, an online meeting and a conversation at the table are recorded, transcribed and summarised the same way, so a candidate’s file reads the same however you spoke to them.

Salary expectation, notice period, required skills. Read out of what was said and written to the record as a proposal you approve.

Decide per kind of conversation what the summary has to cover, so an intake, a second round and a client call do not all come back in the same shape.

Keep the video recording of an online meeting itself, not only its audio.

Pick a template and Simply pulls the data straight out of the ATS: the candidate, the job, the client, whichever fields the template asks for. It comes out in your own branding, with your layout, logo, fonts and colours. Ask the AI in the editor to shorten a section, rewrite an introduction or tailor it to the vacancy, and nothing is copied by hand at any point.

Pick the candidate and the vacancy, and Simply builds the presentation from the record: your template, your branding, and an introduction rewritten to sit against that specific job. Choose the tone, the language, whether experience is rewritten to bullets, and whether the profile goes out anonymised. Export it and the PDF is filed back on the candidate.

Scores a CV out of a hundred on the things that decide whether it gets read: whether achievements are quantified, how much filler is in it, its length, how readable it is, and whether the formatting survives being parsed by the other side. Every finding points at the block it is about and often carries a one-click fix. It runs on the actual text, so it works without an AI key and cannot invent a problem.

Drop in a CV, a vacancy, a job spec or a brief and Simply reads it against your own model. Every value lands in the field it belongs to, including the records underneath it: each job, each education, each skill as its own entry rather than a wall of text in one box. Nothing is saved until you have looked at it.

The candidate keeps their content, you keep your layout, logo, fonts and colours. A CV that arrives as somebody else’s Word file goes out as your document.

Three ways in. Start from one of the ready-made recruitment templates and adjust it. Ask the AI chat to build one for you in plain language. Or draw it yourself in the visual builder: pick what starts it, pick what happens next. No developer needed for any of the three.

Internal means acting inside Simply: updating a record, moving a candidate on, setting a task. External means triggering another tool or channel, like posting to Slack, putting something on a calendar, or updating a record in a connected system.

Google Calendar, Slack, Teams, Outlook, Gmail, LinkedIn, WhatsApp, Zoom, HubSpot, Salesforce, DocuSign and more.

Drag a widget onto the canvas, drop it where you want it, drag its corner to size it. Point it at any object in your data model, pick what it counts and what it groups on, and it draws itself. Build one for yourself, one per recruiter, one for the Monday meeting. No request to us and no report builder to learn.

Line, bar and donut charts, plus single figures for the numbers that need no chart.

Pick the field to group on and the figure to measure, so a dashboard answers your question rather than a default one.

Set a dashboard to a date range, whether that is 1 January to 4 April, last quarter or a single week. Every figure on it then answers for exactly that window.

Take the figures for a period out of Simply, to share with a client or put in front of a management team.

Two people can open the same dashboard and see different totals, because each figure is counted from what that person is allowed to see.

Everything the product does over HTTP, so your own developers can read from Simply and write to it: pull candidates into a report, push vacancies in from your website, wire it into whatever you already run. The documentation is generated from the code, so it cannot describe an endpoint that no longer exists.

MCP is the standard an AI assistant uses to reach a system it does not know. Point your own agent at Simply and it can look things up and prepare changes, without anyone writing an integration for it. It inherits the same rules as everything else here: it only sees what the person behind the key may see, and it can propose but not save.

Keys you mint in Simply so another system, or your own AI agent, can reach your data. Give each one an expiry date and withdraw it whenever you like. Not to be confused with the AI key you connect for model usage, which is yours and billed by your provider.

The other direction from the API: instead of your system asking Simply every few minutes whether anything happened, Simply tells it the moment something does. A candidate is placed, a vacancy is filled, a document is exported, and your own system hears about it straight away. Deliveries are signed so you can tell they came from us, retried if your end is down, and replayable after the fact.

Platform, database and backups run on EU infrastructure we manage ourselves, in the Netherlands. Your candidate data never has to leave the EU, and no model is trained on what you put in.

The models behind search, matching and message labelling run inside our own platform, so those never send your data anywhere. Language-model work is the exception and is stated as such: it runs on European infrastructure, or on your own key with your own provider and region.

Independently certified, with a valid certificate we share on request.

Isolation is structural rather than a setting: a query for another workspace’s data returns nothing, because it never reaches it.

Give someone a role and they get exactly the access that role carries. Change what a role may do and everybody holding it changes with it, so you are never editing permissions person by person.

Not just per screen: who may see which object, which field within it, and which individual records.

Mark a field as sensitive and it stays hidden even from people who may open the record. Salary, date of birth, a citizen service number: visible only to the roles you unlock it for, and the AI cannot read what you cannot.

Some roles see everything in the workspace, others only what is assigned to them. A recruiter working one client does not need to browse the rest, and this is where you decide that.

A code from an authenticator app on top of the password. Enable it per person or require it across the whole workspace.

Sign in with a fingerprint, a face or a device PIN instead of a password. There is no password left to phish, reuse or write on a sticky note.

Your team signs in with the account they already have, so joining and leaving are handled where you already handle them and there is no separate password to manage.

Every write is recorded: who did it, what changed, and when. That includes anything the AI or an automation did, logged under the person who approved it rather than under a system account, so a candidate asking why a decision was made has an answer and an auditor asking who touched a record has one too.

Connect Simply to your own identity provider over OpenID Connect, so access follows your directory rather than a list kept here. The step up from SSO for organisations that run their own identity policy.

Decide how long data stays before it is removed, per kind of record. Set it to what your own policy or your client contract requires rather than to whatever the platform defaults to.

One login, several environments. Handy if you run more than one brand or entity, or if you work for a client in their own workspace: you switch rather than log out.

Invite by email address, pick the role, done. Somebody who already has a Simply account joins your workspace without creating a second one.

Group people the way your agency actually works, by desk, branch or client, and hang access and assignment off that group instead of off individuals.

Your logo and colours on the screen your team signs in on, and on the invitation a new colleague receives.

Sign on your paper rather than ours. We will work through your data processing agreement and your terms instead of asking you to accept a standard set.

Invoicing that fits how your finance team works: per entity, per cost centre, on your own terms and cycle rather than a card charged monthly.

One person who knows your setup, rather than a queue. They are there for the rollout, for the questions after it, and for the conversation when something needs to change.